Transcript packet switching

Information Technology in Theory
By Pelin Aksoy and Laura DeNardis
Chapter 12
Wide Area Networks
Objectives
• Understand the concept of a wide area network
• Identify the main technical components of a wide
area network
• Distinguish between packet switching and circuit
switching
• Understand virtual private networks (VPNs)
Information Technology in Theory
2
Objectives (continued)
• Gain familiarity with the most important
commercial WAN services
• Understand WAN access technologies, including
dedicated lines, xDSL, cable modem access, and
WiMAX
• Identify important network management functions
Information Technology in Theory
3
WAN Background
• Industry convention describes a WAN as a
network that spans a large geographical distance
• The largest example of a WAN is the public
Internet, but many other types of WANs exist
• Wide area networking is sometimes referred to as
enterprise networking
Information Technology in Theory
4
WAN Background (continued)
• Wide area networking developed from the way
businesses first used computer networks to exchange
information internally, beginning in the mid-1970s
• Several architectural features distinguished these
networks from modern WANs
• WANs now support voice, data, and multimedia
information; use open network protocols; and often
are offered over a public network such as the Internet
Information Technology in Theory
5
Packet Switching
• Information sent over the Internet is broken into
small segments called packets
• Each packet contains the actual information
content to be transmitted, as well as the order of
the packet, the sender’s binary address (called the
source address), and the binary address of the
packet’s destination (called the destination
address)
Information Technology in Theory
6
Packet Switching (continued)
• The path that one packet traverses over a network
from source to destination may be different from the
next packet’s path, depending on network congestion
or other conditions
• In this type of networking approach, known as
connectionless packet switching, no dedicated endto-end physical connection is established for the
duration of data transmission
Information Technology in Theory
7
Packet Switching (continued)
• Network devices called routers read the destination
address and determine how to expeditiously route
packets through the networks, based on routing
algorithms that are designed to minimize latency
• Routers are also designed to minimize hops, the
number of times a packet traverses various routers as
it is transmitted over a network
• Once all the packets from a given transmission reach
their destination, they are reassembled in correct
order
Information Technology in Theory
8
Packet Switching (continued)
• The packet-switching approach contrasts with the
circuit-switching approach of the traditional
telephone network
• The circuit-switching approach establishes a physical,
dedicated end-to-end path through the network
between a caller and receiver, and maintains the path
for the entirety of the call
Information Technology in Theory
9
Packet Switching (continued)
Information Technology in Theory
10
Network Protocols
• Network protocols are another important
technical enabler of WANs
• Unlike older types of networks, such as the SNA
and DECnet systems, modern wide area
networking has made the transition to more open
network protocols
• Earlier protocols, such as SNA and DECnet, were
proprietary protocols
Information Technology in Theory
11
Network Protocols (continued)
• The dominant family of network protocols in modern
architectures is TCP/IP
• It is not controlled by a single vendor; it is available
for any manufacturer to use, and it enables universal
access to the Internet
Information Technology in Theory
12
WAN Architectural Components
• A router is an intelligent switching device that
determines how to direct (or route) a packet across
a network, based on the packet’s destination
address and network conditions
• An enormous, interconnected web of thousands of
routers makes up the backbone of the Internet
Information Technology in Theory
13
WAN Architectural Components
(continued)
• The router “looks up” information in a routing table
• These routing tables are constantly changing and
automatically updated as routers probe their network
environments, exchange information with other
routers, and dynamically update their routing tables
Information Technology in Theory
14
WAN Architectural Components
(continued)
Information Technology in Theory
15
WAN Architectural Components
(continued)
• Customers who access the Internet or another type of
WAN are connected to an edge router
• A device called a CSU/DSU (Channel Service
Unit/Data Service Unit) is an important network
component between the edge router and the dedicated
transmission line
• The equipment at a customer location that serves as
the demarcation point between a local network and a
service provider’s WAN is often called customer
premises equipment (CPE)
Information Technology in Theory
16
WAN Architectural Components
(continued)
Information Technology in Theory
17
Private Networks
• If a company such as a bank wants a dedicated
transmission path between its branches for private
digital communication lines, the company has the
option of leasing dedicated private lines from a
network provider
• A private line is not shared with other customers,
so it can guarantee performance and availability to
some degree
Information Technology in Theory
18
Private Networks (continued)
• These dedicated lines may run terrestrially over
fiber-optic cables or high-grade twisted pair, via
undersea fiber-optic cables, or even via satellite
for international communications or remote areas
that do not have adequate terrestrial
telecommunications
• The primary disadvantage of a private network is
cost, because a single enterprise bears the entire
cost of the lines
Information Technology in Theory
19
Private Networks (continued)
Information Technology in Theory
20
Internet Services
• Many businesses use the infrastructure of the
public Internet to communicate internally and with
customers and suppliers
• Using the Internet and other shared public WANs
is much more cost effective than using dedicated
private networks
• Virtual private networks (VPNs) are a response
to the performance and security concerns of public
networks
Information Technology in Theory
21
Internet Services (continued)
Information Technology in Theory
22
Internet Services (continued)
• Privacy and security are achieved through the
application of security measures and a technique
called tunneling
• VPN services offer quality-of-service (QoS)
guarantees to their customers to cover such factors as
latency of packets, dropped packets that cannot be
routed to their destination, and network availability
• VPN customers usually obtain service level
agreements (SLAs) from the VPN provider
Information Technology in Theory
23
Frame Relay Service
• The frame relay network is owned and operated by
the service provider but is used by the customer
• This service uses packets, called frames, with a
variable number of bits that are switched
throughout the WAN until they reach their
destination
• The frame relay standard uses a different format
from other WAN standards
Information Technology in Theory
24
Frame Relay Service (continued)
• A software-defined (virtual) path is set between two
devices on the network, and the devices are expected
to exchange information
– Switched virtual circuits (SVCs)
– Permanent virtual circuits (PVCs)
• Historically, frame relay originated as an
improvement over an older WAN technology known
as X.25
Information Technology in Theory
25
Asynchronous Transfer Mode
• WAN service providers also offer Asynchronous
Transfer Mode (ATM), a network alternative that
formats information into fixed-length packets
• These packets are normally called cells in the
context of ATM
• ATM cells have a total length of 53 bytes
• ATM is a connection-oriented WAN approach
• ATM’s fixed transmission delays, virtual circuits,
and fixed cell size are beneficial for low latency
applications
Information Technology in Theory
26
Multiprotocol Label Switching
• A more practical type of WAN service, called
multiprotocol label switching (MPLS), is
designed to simultaneously support many types of
WAN traffic
• MPLS service can handle variable-length packets,
in contrast with ATM, which transmits fixedlength cells
• One major difference between MPLS and
traditional IP packet switching is that MPLS is
connection oriented
Information Technology in Theory
27
Multiprotocol Label Switching
(continued)
• Traditional packet switching is often connectionless,
meaning that a dedicated end-to-end connection is not
established for the duration of transmission
• Unlike this approach, MPLS routes packets along
preconfigured paths
Information Technology in Theory
28
WAN Access Alternatives
• The access mechanism selected depends on
several variables
– Bandwidth
– Mobility
– Security
– Availability
Information Technology in Theory
29
Leased Private Lines
• Businesses and other enterprises that have more
than a dozen employees and that need WAN
access usually lease a dedicated private line to a
commercial WAN service
Information Technology in Theory
30
Leased Private Lines (continued)
• The line is rented on a monthly basis from a
telecommunications service provider, and comes in
the following transmission speeds:
–
–
–
–
–
–
768 Kbps: Sub-T1 link (also called a fractional T1 line)
T-1 link: 1.544 Mbps (also called a dedicated T1 line)
T-3 link: 45 Mbps
OC-3: 155 Mbps
OC-12: 622 Mbps
OC-48: 2.488 Gbps
Information Technology in Theory
31
Leased Private Lines (continued)
Information Technology in Theory
32
Digital Subscriber Line
• DSL is a WAN access alternative that connects a
user’s DSL modem to the twisted pair cables installed
as part of the traditional telephone network
• Many types of DSL technology are available:
– Symmetric digital subscriber line (SDSL)
– Asymmetric digital subscriber line (ADSL)
– High data rate digital subscriber line (HDSL)
Information Technology in Theory
33
Digital Subscriber Line (continued)
• DSL quality depends on the distance between the user
site and the telephone company (the “central office”)
that houses the DSL termination equipment
• This equipment is known as the DSLAM, or DSL
access multiplexer
• DSL service can offer a transmission rate of up to 7.1
Mbps downstream and approximately 768 Kbps
upstream
Information Technology in Theory
34
Digital Subscriber Line (continued)
Information Technology in Theory
35
Digital Subscriber Line (continued)
Information Technology in Theory
36
Digital Subscriber Line (continued)
Information Technology in Theory
37
Cable Modem Access
• Internet access via cable offers transmission rates
in the megabit per second range—approximately
30 to 50 Mbps downstream and 5 Mbps upstream
in some areas
• Cable WAN access is geared primarily to
residential customers who already subscribe to
cable television service and is offered as part of a
package with cable television, high-speed Internet
access, and Voice over IP service
Information Technology in Theory
38
Cable Modem Access (continued)
Information Technology in Theory
39
WiMAX
• High-speed wireless broadband access to the
Internet is commonplace over small geographical
areas through technologies such as Wi-Fi
• A significant technological requirement is for
products and services to provide the same type of
wireless WAN access over greater distances, such
as across a city using a metropolitan area network
Information Technology in Theory
40
WiMAX (continued)
• WiMAX (Worldwide Interoperability for Microwave
Access)
• Another name for IEEE 802.16
• A formal set of networking standards for wireless
metropolitan area networks
• Developed by the Institute of Electrical and
Electronics Engineers (IEEE)
Information Technology in Theory
41
WiMAX (continued)
• WiMAX networks would consist of WiMAX
antennas transmitting to residences and businesses
with a WiMAX receiver, to laptops with WiMAX
cards, or to Wi-Fi hot spots.
• WiMAX provides a theoretical maximum uplink and
downlink speed of approximately 70 Mbps, although
this rate can be achieved only over short ranges
• Performance varies depending on distance, spectrum
allocated, and number of concurrent users sharing
bandwidth
Information Technology in Theory
42
WiMAX (continued)
Information Technology in Theory
43
Network Management Systems
• Businesses, universities, and even individual users
require local and wide area network services that are
always available and that perform reliably
• To achieve these requirements, users must provide or
outsource network management activities, which fall
into four categories:
– Configuration management
– Performance management
– Fault management
– Security management
Information Technology in Theory
44
Configuration Management
• The function of configuration management is to
track and manage all the hardware and software
associated with the computer network, and to
manage any changes that occur to these
architectural elements
Information Technology in Theory
45
Configuration Management
(continued)
Information Technology in Theory
46
Performance Management
• Another function that network managers must
provide is performance management, which
ensures that the network is performing adequately
for the applications and users it supports
Information Technology in Theory
47
Fault Management
• One of the most important functions of network
management is fault management—solving a
network outage or performance problem when it
occurs
Information Technology in Theory
48
Security Management
• Network security is a vital management function that
is often handled by a dedicated security team
• Some of these security management functions
include network access control, user authentication,
firewall management, and critical infrastructure
protection
• Some security threats to enterprise networks include
worms and viruses, spam, unauthorized access
attempts, and data interception
Information Technology in Theory
49
Summary
• A WAN is a network that spans a large geographical
distance and can transmit voice, data, and multimedia
information
• The Internet is the most prominent example of a WAN
• Most WAN services are run over a public network
rather than private lines due to cost concerns
• Most WANs, including the Internet, are based on a
network approach known as packet switching, which
breaks information into small segments called packets
prior to transmission
– These packets are sent over a network, possibly routed along
different paths, and reassembled at their destination
Information Technology in Theory
50
Summary (continued)
• A virtual private network (VPN) runs over a public
network like the Internet, but it emulates a private
network’s higher performance and security by applying
security measures like encryption and by offering quality
of service (QoS) guarantees
• WANs require network protocols, the standard rules that
allow information to be exchanged over the network
• Important types of WANs are Internet services, frame
relay, Asynchronous Transfer Mode, and multiprotocol
label switching
Information Technology in Theory
51
Summary (continued)
• Most businesses and large institutions access WANs via
leased lines, while smaller businesses and individual users
access WANs via wireless technologies, DSL, and cable
modems
• Ensuring adequate WAN performance and reliability
requires network management functions, including
configuration management, performance management,
fault management, and security management
Information Technology in Theory
52